Suboptimal Criterion Learning in Static and Dynamic Environments

Elyse H. Norton, Stephen M. Fleming, Nathaniel D. Daw, Michael Landy

Research output: Contribution to journalArticle

Abstract

Humans often make decisions based on uncertain sensory information. Signal detection theory (SDT) describes detection and discrimination decisions as a comparison of stimulus “strength” to a fixed decision criterion. However, recent research suggests that current responses depend on the recent history of stimuli and previous responses, suggesting that the decision criterion is updated trial-by-trial. The mechanisms underpinning criterion setting remain unknown. Here, we examine how observers learn to set a decision criterion in an orientation-discrimination task under both static and dynamic conditions. To investigate mechanisms underlying trial-by-trial criterion placement, we introduce a novel task in which participants explicitly set the criterion, and compare it to a more traditional discrimination task, allowing us to model this explicit indication of criterion dynamics. In each task, stimuli were ellipses with principal orientations drawn from two categories: Gaussian distributions with different means and equal variance. In the covert-criterion task, observers categorized a displayed ellipse. In the overt-criterion task, observers adjusted the orientation of a line that served as the discrimination criterion for a subsequently presented ellipse. We compared performance to the ideal Bayesian learner and several suboptimal models that varied in both computational and memory demands. Under static and dynamic conditions, we found that, in both tasks, observers used suboptimal learning rules. In most conditions, a model in which the recent history of past samples determines a belief about category means fit the data best for most observers and on average. Our results reveal dynamic adjustment of discrimination criterion, even after prolonged training, and indicate how decision criteria are updated over time.

Original languageEnglish (US)
Article numbere1005304
JournalPLoS Computational Biology
Volume13
Issue number1
DOIs
StatePublished - Jan 1 2017

Fingerprint

Dynamic Environment
learning
Learning
ellipse
Discrimination
Observer
history
History
Signal detection
Gaussian distribution
Social Adjustment
Normal Distribution
Ellipse
Data storage equipment
Discrimination (Psychology)
decision
Research
trial
Rule Learning
Signal Detection

ASJC Scopus subject areas

  • Ecology, Evolution, Behavior and Systematics
  • Modeling and Simulation
  • Ecology
  • Molecular Biology
  • Genetics
  • Cellular and Molecular Neuroscience
  • Computational Theory and Mathematics

Cite this

Suboptimal Criterion Learning in Static and Dynamic Environments. / Norton, Elyse H.; Fleming, Stephen M.; Daw, Nathaniel D.; Landy, Michael.

In: PLoS Computational Biology, Vol. 13, No. 1, e1005304, 01.01.2017.

Research output: Contribution to journalArticle

Norton, Elyse H. ; Fleming, Stephen M. ; Daw, Nathaniel D. ; Landy, Michael. / Suboptimal Criterion Learning in Static and Dynamic Environments. In: PLoS Computational Biology. 2017 ; Vol. 13, No. 1.
@article{5f9f6c3a7e5a437f9073ddd3e7aaf4a2,
title = "Suboptimal Criterion Learning in Static and Dynamic Environments",
abstract = "Humans often make decisions based on uncertain sensory information. Signal detection theory (SDT) describes detection and discrimination decisions as a comparison of stimulus “strength” to a fixed decision criterion. However, recent research suggests that current responses depend on the recent history of stimuli and previous responses, suggesting that the decision criterion is updated trial-by-trial. The mechanisms underpinning criterion setting remain unknown. Here, we examine how observers learn to set a decision criterion in an orientation-discrimination task under both static and dynamic conditions. To investigate mechanisms underlying trial-by-trial criterion placement, we introduce a novel task in which participants explicitly set the criterion, and compare it to a more traditional discrimination task, allowing us to model this explicit indication of criterion dynamics. In each task, stimuli were ellipses with principal orientations drawn from two categories: Gaussian distributions with different means and equal variance. In the covert-criterion task, observers categorized a displayed ellipse. In the overt-criterion task, observers adjusted the orientation of a line that served as the discrimination criterion for a subsequently presented ellipse. We compared performance to the ideal Bayesian learner and several suboptimal models that varied in both computational and memory demands. Under static and dynamic conditions, we found that, in both tasks, observers used suboptimal learning rules. In most conditions, a model in which the recent history of past samples determines a belief about category means fit the data best for most observers and on average. Our results reveal dynamic adjustment of discrimination criterion, even after prolonged training, and indicate how decision criteria are updated over time.",
author = "Norton, {Elyse H.} and Fleming, {Stephen M.} and Daw, {Nathaniel D.} and Michael Landy",
year = "2017",
month = "1",
day = "1",
doi = "10.1371/journal.pcbi.1005304",
language = "English (US)",
volume = "13",
journal = "PLoS Computational Biology",
issn = "1553-734X",
publisher = "Public Library of Science",
number = "1",

}

TY - JOUR

T1 - Suboptimal Criterion Learning in Static and Dynamic Environments

AU - Norton, Elyse H.

AU - Fleming, Stephen M.

AU - Daw, Nathaniel D.

AU - Landy, Michael

PY - 2017/1/1

Y1 - 2017/1/1

N2 - Humans often make decisions based on uncertain sensory information. Signal detection theory (SDT) describes detection and discrimination decisions as a comparison of stimulus “strength” to a fixed decision criterion. However, recent research suggests that current responses depend on the recent history of stimuli and previous responses, suggesting that the decision criterion is updated trial-by-trial. The mechanisms underpinning criterion setting remain unknown. Here, we examine how observers learn to set a decision criterion in an orientation-discrimination task under both static and dynamic conditions. To investigate mechanisms underlying trial-by-trial criterion placement, we introduce a novel task in which participants explicitly set the criterion, and compare it to a more traditional discrimination task, allowing us to model this explicit indication of criterion dynamics. In each task, stimuli were ellipses with principal orientations drawn from two categories: Gaussian distributions with different means and equal variance. In the covert-criterion task, observers categorized a displayed ellipse. In the overt-criterion task, observers adjusted the orientation of a line that served as the discrimination criterion for a subsequently presented ellipse. We compared performance to the ideal Bayesian learner and several suboptimal models that varied in both computational and memory demands. Under static and dynamic conditions, we found that, in both tasks, observers used suboptimal learning rules. In most conditions, a model in which the recent history of past samples determines a belief about category means fit the data best for most observers and on average. Our results reveal dynamic adjustment of discrimination criterion, even after prolonged training, and indicate how decision criteria are updated over time.

AB - Humans often make decisions based on uncertain sensory information. Signal detection theory (SDT) describes detection and discrimination decisions as a comparison of stimulus “strength” to a fixed decision criterion. However, recent research suggests that current responses depend on the recent history of stimuli and previous responses, suggesting that the decision criterion is updated trial-by-trial. The mechanisms underpinning criterion setting remain unknown. Here, we examine how observers learn to set a decision criterion in an orientation-discrimination task under both static and dynamic conditions. To investigate mechanisms underlying trial-by-trial criterion placement, we introduce a novel task in which participants explicitly set the criterion, and compare it to a more traditional discrimination task, allowing us to model this explicit indication of criterion dynamics. In each task, stimuli were ellipses with principal orientations drawn from two categories: Gaussian distributions with different means and equal variance. In the covert-criterion task, observers categorized a displayed ellipse. In the overt-criterion task, observers adjusted the orientation of a line that served as the discrimination criterion for a subsequently presented ellipse. We compared performance to the ideal Bayesian learner and several suboptimal models that varied in both computational and memory demands. Under static and dynamic conditions, we found that, in both tasks, observers used suboptimal learning rules. In most conditions, a model in which the recent history of past samples determines a belief about category means fit the data best for most observers and on average. Our results reveal dynamic adjustment of discrimination criterion, even after prolonged training, and indicate how decision criteria are updated over time.

UR - http://www.scopus.com/inward/record.url?scp=85011422144&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85011422144&partnerID=8YFLogxK

U2 - 10.1371/journal.pcbi.1005304

DO - 10.1371/journal.pcbi.1005304

M3 - Article

C2 - 28046006

AN - SCOPUS:85011422144

VL - 13

JO - PLoS Computational Biology

JF - PLoS Computational Biology

SN - 1553-734X

IS - 1

M1 - e1005304

ER -